This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
[patch] amdfam10 costs for target cost model
- From: "Sjodin, Jan" <Jan dot Sjodin at amd dot com>
- To: gcc-patches at gcc dot gnu dot org
- Date: Sun, 9 Sep 2007 14:07:22 -0500
- Subject: [patch] amdfam10 costs for target cost model
This patch is dependent on patch
http://gcc.gnu.org/ml/gcc-patches/2007-09/msg00547.html
Since the main patch is likely to be changed, these cost may also
change. The plan is to test the polyhedron benchmark suite after the
cost model is stable and these costs can be evaluated again.
2007-09-09 Jan Sjodin <jan.sjodin@amd.com>
* config/i386/i386.c (amdfam10cost): Tuned costs for cost model
*** config/i386/i386.c 2007-09-09 10:20:53.000000000 -0500
--- config/i386/i386.c 2007-09-09 10:15:33.000000000 -0500
*************** struct processor_costs amdfam10_cost = {
*** 784,798 ****
{{libcall, {{8, loop}, {24, unrolled_loop},
{2048, rep_prefix_4_byte}, {-1, libcall}}},
{libcall, {{48, unrolled_loop}, {8192, rep_prefix_8_byte}, {-1,
libcall}}}},
! 1, /* scalar_stmt_cost. */
! 1, /* scalar load_cost. */
! 1, /* scalar_store_cost. */
! 1, /* vec_stmt_cost. */
! 1, /* vec_to_scalar_cost. */
! 1, /* scalar_to_vec_cost. */
! 1, /* vec_align_load_cost. */
! 2, /* vec_unalign_load_cost. */
! 1, /* vec_store_cost. */
};
static const
--- 784,798 ----
{{libcall, {{8, loop}, {24, unrolled_loop},
{2048, rep_prefix_4_byte}, {-1, libcall}}},
{libcall, {{48, unrolled_loop}, {8192, rep_prefix_8_byte}, {-1,
libcall}}}},
! 4, /* scalar_stmt_cost. */
! 2, /* scalar load_cost. */
! 2, /* scalar_store_cost. */
! 6, /* vec_stmt_cost. */
! 0, /* vec_to_scalar_cost. */
! 2, /* scalar_to_vec_cost. */
! 2, /* vec_align_load_cost. */
! 3, /* vec_unalign_load_cost. */
! 2, /* vec_store_cost. */
};